Within the field of computing, many scenarios involve a presentation of messages in a message queue, such as a set of email messages presented in an email client, a set of private messages sent to a user in a web forum, or a set of chat messages received by various other users in a chat environment. In such scenarios, some messages may present undesired content, such as bulk unsolicited messages (“spam”); advertisements for goods or services that are illegal, illegitimate, or simply irrelevant to the user; objectionable material; malicious content, such as attachments or links to maliciously crafted software; and “phishing” attempts that impersonate a trusted individual in order to exploit the trust of the user.
Various techniques may utilized to evaluate the messages in order to identify whether the such messages are desirable or undesirable, such as a verification of the sender of the message, a semantic evaluation of the content of the message, and an evaluation of any links or software included in the message for malicious content. Messages identified as unwanted may be automatically deleted, partitioned (e.g., moving such messages to a “spam” folder), or otherwise filtered out of the message queue. Such filtering techniques are carefully formulated to reduce the incidence of false positives, wherein a legitimate message is incorrectly identified as an unwanted message and filtered out of the message queue, thereby resulting in a missed message that the user may have desired to have included in the message queue.